Connect public, paid and private patent data with Google Patents Public Datasets

Method for ensuring data coherency for redundat storage by transferring copies of access requests from on-line storage to back-up storage

Download PDF

Info

Publication number
US5845328A
US5845328A US08766455 US76645596A US5845328A US 5845328 A US5845328 A US 5845328A US 08766455 US08766455 US 08766455 US 76645596 A US76645596 A US 76645596A US 5845328 A US5845328 A US 5845328A
Authority
US
Grant status
Grant
Patent type
Prior art keywords
storage
unit
line
command
back
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
US08766455
Inventor
Yuzuru Maya
Soichi Isono
Akira Ohtsuji
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hitachi Ltd
Original Assignee
Hitachi Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Grant date

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRICAL DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/16Error detection or correction of the data by redundancy in hardware
    • G06F11/20Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
    • G06F11/2053Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where persistent mass storage functionality or persistent mass storage control functionality is redundant
    • G06F11/2056Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where persistent mass storage functionality or persistent mass storage control functionality is redundant by mirroring
    • G06F11/2071Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where persistent mass storage functionality or persistent mass storage control functionality is redundant by mirroring using a plurality of controllers

Abstract

A command is issued from a processing unit to an on-line storage unit and transferred from the on-line storage unit to a back-up storage unit. The received commands are processed in the respective storage units. An overhead of the on-line storage unit is reduced in a hot stand-by system. Storage contents of the on-line storage unit and the back-up storage unit are made coincident in a structure having a plurality of processing units, where the on-line storage unit and the back-up storage unit are connected through a channel device.

Description

BACKGROUND OF THE INVENTION

The present invention relates to a method for making storage contents of duplex storage unit including an on-line storage unit and a back-up storage unit coincident, and more particularly to a method for making storage contents of the duplex storage unit coincident, which assures a reception order of commands and eliminates the need to transfer a large volume of checkpoint data by transferring commands received by the on-line storage unit from a processing unit to the back-up storage unit.

In a prior art computer system having duplex storage unit connected to a plurality of processing units (central processing complexity), when a fault occurs in one of the duplex storage unit, the fault is notified to the plurality of processing units, the other storage unit is selected and the fault is recovered by making use of a recovery function of a parallel transaction server.

However, when such duplex storage unit is applied to a structure of a hot stand-by system, an amount of checkpoint data to be transferred from an on-line system to a back-up system increases due to the increase of the number of processing units and the enhancement of a processing capability, and an overhead of the on-line storage unit to transfer the data increases.

Namely, in a conventional hot stand-by system, an on-line processing unit transmits checkpoint data (checkpoint dump) to a back-up processing unit at a constant time interval. When a fault occurs in the on-line unit, the stand-by processing unit succeeds the process of the on-line unit by the checkpoint data received from the on-line system. Accordingly, when this system is applied to the computer system having the duplex storage unit, a volume of checkpoint data to be transferred from the on-line system to the back-up system increases by the increase of the number of processing units and the enhancement of the processing capability and the overhead of the on-line storage unit to transfer the data increases.

On the other hand, a hot stand-by system in which the storage contents are made coincident by assuring the simultaneous reception of commands (messages) by the on-line storage unit and the back-up storage unit without transferring the checkpoint data has been known. In this system, the on-line storage unit and the back-up storage unit are interconnected through a local area network (LAN) or a line switching device and a transmitting station of the commands occupies the LAN or the line selector device to transmit the commands in such a manner that reception orders of commands in the on-line system and the back-up system coincide.

However, when a plurality of processing units and the on-line storage unit and the back-up storage unit are connected through a channel device, the plurality of processing units may separately issue commands to the on-line storage unit and the back-up storage unit. Thus, the reception orders of the on-line storage unit and the back-up storage unit do not coincide and the storage contents of the on-line storage unit and the back-up storage unit do not coincide.

U.S. Pat. No. 5,537,574 discloses a structure to maintain data called a cache structure in a storage unit. U.S. Pat. No. 5,317,739 discloses a structure to maintain data called a list structure in a storage unit.

SUMMARY OF THE INVENTION

It is an object of the present invention to provide a method for making storage contents of duplex storage unit coincident, which can reduce an overhead of an on-line storage unit when the duplex storage unit is applied to a computer structure of a hot-stand-by system.

It is another object of the present invention to provide a method for making storage contents of an on-line storage unit and a back-up storage unit even in a structure having a plurality of processing units, the on-line storage unit and the back-up storage unit connected through a channel device.

In order to achieve the above object, the present invention is characterized by that the processing unit issues commands to only the on-line storage unit and causes the on-line storage unit to transfer the command to the back-up storage unit, and the processes for the received commands are executed by the respective storage units.

When the processing unit accesses the storage unit, it issues the command to only the on-line storage unit. When the on-line storage unit receives the command, it transfers the command to the back-up storage unit and executes the process for the received command and returns a process result to the processing unit.

On the other hand, the back-up storage unit executes the process for the command received from the on-line storage unit.

Accordingly, the reception orders of the commands in the on-line storage unit and the back-up storage unit always coincide, and the storage contents of both storage units coincide even if the on-line storage unit and the back-up storage unit are connected through the channel device.

Further, since the on-line storage unit transmits commands which are less in volume than the checkpoint data to the back-up storage unit in place of the checkpoint data, the overhead of the on-line storage unit is reduced.

The storage unit includes a cache structure to provide a data cache function and a list structure to provide a message communication function and a data lock function.

When the processing unit having such structures accesses the storage unit, the on-line storage unit executes the process for the command received from the processing unit and returns the process result to the processing unit. In this case, the on-line storage unit decodes the received command, and if the received command is one to update the cache structure or the list structure, it transmits the command to the back-up storage unit. The back-up storage unit executes the process for the received command and returns a response to the on-line storage unit. Finally, the on-line storage unit returns the response to the processing unit.

However, if the received command is not one to update the structure, the on-line storage unit does not transmit the received command to the back-up storage unit. Only the on-line storage unit executes the process for the command and returns a response to the processing unit.

As a result, the on-line storage unit and the back-up storage unit execute the same commands in the same order so that the storage contents of the on-line storage unit and the back-up storage unit coincide.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 shows a system configuration of hardware-of a parallel computer system in accordance with the present invention,

FIG. 2 shows a software configuration of a duplex storage unit,

FIG. 3 shows a cache structure in an on-line storage unit and a back-up storage unit,

FIG. 4 shows a list structure in the on-line storage unit and the back-up storage unit,

FIG. 5 shows classification of execution results of commands to the on-line storage unit and the back-up storage unit,

FIG. 6 shows a flow chart of a process in the storage unit for a command received from a processing unit,

FIG. 7 shows a flow chart which continues from FIG. 6,

FIG. 8 illustrates a process at time-out,

FIG. 9 illustrates a content of a resumption table at the time-out,

FIGS. 10A, 10B and 10C illustrate response information at the time-out,

FIG. 11 illustrates retry of a command, and

FIG. 12 shows a flow chart of a coincidence control process of a status transition command.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

Embodiments of the present invention are now explained in detail with reference to the drawings.

FIG. 1 shows a system configuration of an embodiment of a parallel computer system in accordance with the present invention and it comprises an on-line storage unit 10, a back-up storage unit 11 and three processing units 20˜22.

Hardware configurations of the storage units 10 and 11 and the processing units 20˜22 are identical and comprise command processors 10-1˜22-1, memories 10-2˜22-2, IOP 10-3˜22-3, system supervisor devices 10-4˜22-4 and input/output channels 10-6˜22-6, respectively.

Each of the processing units 20˜22 is connected to the two storage units 10 and 11 by the input/output channels 20-6˜22-6. In other words, the two storage units 10 and 11 can be accessed from any one of the processing units 20˜22.

System supervisor devices 10-4˜11-4 of the two storage units 10 and 11 are connected by a control message path 3 through which a message to reset the storage unit of the partner system is communicated.

The processing units 20˜22 share a disk 4.

The storage units 10 and 11 include a cache structure to provide a data cache function of the processing units 20˜22 and a list structure to provide a message communication function among the processing units 20˜22 and an exclusive control function of data.

FIG. 2 shows a software configuration of the storage units 10 and 11. The on-line storage unit 10 comprises a supervisor 10-10, a command processor 10-11 and a coincidence control processor 10-12. It further comprises a buffer 10-13 for storing a received command and a resumption table 10-14 which is used when a timer to monitor a command process is timed out.

The command processor 10-11 executes the command and the coincidence control processor 10-12 checks if execution results of the command in the on-line storage unit 10 and the back-up storage unit 11 coincide.

The software of the back-up storage unit 11 is of identical configuration to the software of the on-line storage unit 10 and comprises a supervisor 11-10, a command processor 11-11 and a coincidence control processor 11-12. It further comprises a buffer 11-13 for storing the received command and a resumption table 11-14 used at the time-out.

An operation of the above configuration is described below.

When the processing units 20˜22 access the on-line and back-up storage units 10 and 11, the processing units 20˜22 transmit corresponding commands to the on-line storage unit 10.

When the command processor 10-11 of the on-line storage unit 10 receives the command from any one of the processing units 20˜22, it executes a process for the received command to access the cache structure 30-0 or the list structure 40-0 in the storage unit 10 (step 51).

If the command is one to update the content of the storage units 10 and 11 (the cache structure or the list structure), the supervisor 10-10 of the on-line storage unit 10 transmits that command to the back-up storage unit 11 (step 52). The command is transmitted through the communication path 5 between the storage units 10 and 11.

The command processor 11-11 of the back-up storage unit 11 executes a process for the received command, that is, the updating process for the cache structure 30-1 or the list structure 40-1 (step 53) and the supervisor 11-10 transmits back a response massage to the on-line storage unit 10 (step 54). The response message is transmitted through the communication path 5 between the storage units 10 and 11.

The coincidence control processor 10-12 of the on-line storage unit 10 checks if the execution results for the command in both storage units 10 and 11 coincide or not (step 55), and returns the result to the processing units 20˜22 (step 56). Thereafter, the processing units execute process in accordance with the received result.

The on-line storage unit 10, however, does not transmit the commands received from the processing units 20˜22 to the back-up storage unit 11 if the command is not one to update the cache structure or the list structure.

Accordingly, if the command received from the processing units 20˜22 is not the update command for the cache structure or the list structure, only the on-line storage unit 10 executes the process for the command and transmits back the read data to the processing units 20˜22 (step 57).

If a status transition occurs in the cache structure 30-0 or the list structure 40-0 of the on-line storage unit 10 (step 60), the on-line storage unit 10 issues a status transition command to notify the status transition to the processing units 20˜22 (step 61).

On the other hand, if the status transition occurs in the cache structure 30-1 or the list structure 40-1 of the back-up storage unit 11 (step 62), the back-up storage unit 11 issues a status transition command to the on-line back-up 10 (step 63). The status transition command is transmitted through the communication path 5 between the storage units 10 and 11.

The on-line storage unit 10 determines whether the same status transition command has been issued in the storage units 10 and 11 within a predetermined period by the coincidence control processor 10-12 (step 64). The subsequent steps will be described in detail with reference to FIG. 12.

FIG. 3 shows detail of the cache structure 30-1

As described in U.S. Pat. No. 5,537,574 "Sysplex Shared Data Coherency Method", the cache structure comprises cache structure control information 30-3, local cache structure control information 30-4, castout class control information 30-5, a directory entry 30-7 and cache structure data 30-8.

The cache structure control information 30-3 is control information relating to the cache structure such as a structure size and the number of castout classes. The local cache control information 30-4 is information which is effective when a local cache (a cache in the processing unit) is allocated to an SES cache (a cache in the storage unit).

The castout class control information 30-5 is information for a castout process to store updated data in the disk 4.

The directory entry 30-7 includes a cache name 30-10 and a data area pointer 30-11. Data is identified by the cache name 30-10 and a storage location of the cache structure data 30-8 is specified by the data area pointer 30-11.

The number of directory entries 30-7 has an upper limit, and when the upper limit is reached, a message of memory shortage is notified to the processing units 20˜22.

FIG. 4 shows detail of the list structure.

As described in U.S. Pat. No. 5,317,739 "Method and Apparatus for Coupling Data Processing Systems", the list structure comprises list structure control information 40-3, user control information 40-4, a lock table 40-5, list control information 40-6, list entry control information 40-10, list structure data 40-18 and adjunct list structure data 40-19. The next list entry control information 40-20 is associated with the list entry control information 40-10 by a pointer.

The list structure control information 40-3 is control information relating to the list structure such as a structure size and the number of lists. The user control information 40-4 if user-defined information. The lock table 40-5 is a table to exclusively control the reading and the writing of the list entry control information 40-10.

The list control information 40-6 is information indicating a count of the list entries and a maximum value thereof. The list entry control information 40-10 and 40-20 comprise list entry identifiers 40-11 and 40-21 for identifying the list structure data 40-18 and 40-28 and the adjunct list structure data 40-19 and 40-29, pointers (data area pointers) 40-12 and 40-22 to the data 40-18 and 40-58, pointers (adjunct data pointers) 40-13 and 40-23 to the adjunct data 40-19 and 40-29, and pointers (next list entry pointers) 40-14 and 40-24 to the next list entry information.

The numbers of the list entry control information 40-10 and 40-2 have upper limits, and when the upper limit is reached, a message of memory shortage is notified to the processing units 20˜22.

FIG. 5 shows an execution result for the command.

The command execution result includes four classes, normal completion 80, time-out 81, memory shortage 82 and structure fault 83.

The normal completion 80 results in when the command is correctly executed. The status of the storage unit is normal and the execution result of the command in the on-line storage unit 10 and the execution result of the command in the back-up storage unit 11 necessarily coincide.

As for the time-out 81, the status of the storage units 10 and 11 are normal but the storage units 10 and 11 return responses to the processing units within a predetermined time period because the amount of execution process is large or the waiting time is long. The execution results of the command in the on-line storage unit 10 and the back-up storage unit 11 do not always coincide if the resumption point is included. One of the storage units 10 and 11 may normally complete.

The memory shortage 82 indicates the non-execution of the write process for the write command because of lack of empty entry when the directory entry information 30-7 or the list entry control information 40-10 and 40-20 are written. The execution results of the command in the on-line storage unit 10 and the back-up storage unit 11 always coincide.

The fault 83 indicates that the access to the memories 10-2 and 11-2 of the storage units 10 and 11 is not attained.

FIGS. 6 and 7 show flow charts of a process for the command received by the on-line storage unit 10.

When the on-line storage unit 10 receives the command from the processing units 20˜22, it executes the process for the received command (step 100).

Then, the on-line storage unit 10 transmits the received command to the back-up storage unit 11 (step 101). Then, the execution result of the command by the on-line storage unit 10 is determined (step 102).

When the execution result of the on-line storage unit 10 indicates the normal completion or the memory shortage, the on-line storage unit 10 transmits the execution result of the command to the processing units 20˜22 without waiting the execution result of the command by the back-up storage unit 11 (step 103). The execution result of the command by the back-up storage unit 11 is monitored (step 104).

When the execution result of the command by the on-line storage unit 10 indicates the memory shortage (step 105), whether both the execution result of the on-line storage unit 10 and the execution result of the back-up storage unit 11 indicate the memory shortage or not is determined (step 112), and if both indicate the memory shortage, the on-line storage unit 10 checks if a command is stored in the buffer or not (step 120). If a command is stored in the buffer 10-13, the process returns to the step 100 to continue the processing of the next command. If no command is stored in the buffer 10-13, the reception of a command is monitored (step 122).

When the execution result of the command by the back-up storage unit 11 does indicate the memory shortage, it means that the execution result of the command by the on-line storage unit 10 and the execution result of the command by the back-up storage unit 11 do not coincide and it is determined that a fault has occurred in the back-up storage unit 11 and the back-up storage unit 11 is reset (step 114).

When the execution result of the command by the on-line storage unit 10 indicates the normal completion (step 105), whether the execution result of the command by the back-up storage unit 11 indicates the normal completion or not is determined (step 106).

When the execution result of the command by the back-up storage unit 11 indicates the normal completion, the processing of the next command is executed (step 120).

When the execution result of the command by the back-up storage unit 11 does not indicate the normal completion, whether the execution result of the command by the back-up storage unit 11 indicates the time-out or not is determined (step 107).

If the execution result of the command by the back-up storage unit 11 indicates the time-out, the back-up storage unit 11 execute the command until the last (step 111). If the execution result of the command by the back-up storage unit 11 indicates the memory shortage or the fault, the back-up storage unit 11 is reset (step 110).

On the other hand, in the step 102, if the execution result of the command by the on-line storage unit 10 indicates the time-out, the on-line storage unit 10 waits the execution result of the back-up storage unit 11 (step 130). Then, the execution result of the back-up storage unit 11 is determined (step 131). If the execution result of the back-up storage unit 11 indicates the normal completion or the time-out, a resumption point is set in the resumption table 10-14 (step 132). Then, the on-line storage unit 10 transmits the execution result of its own command to the processing units 20˜22 (step 133).

In the step 131, if the execution result of the back-up storage unit 11 indicates the memory shortage or the fault, it is determined that a fault has occurred in the back-up storage unit 11 and the back-up storage unit 11 is reset (step 134). Then, the processing of the next command (step 120) is executed.

In the step 102, if the execution result of the on-line storage unit 10 indicates the fault, the on-line storage unit 10 waits the execution result of the back-up storage unit 11 (step 140). Then, the on-line storage unit 10 transmits the execution result of the command by the back-up storage unit 11 to the processing units 20˜22 (step 141). Then, the back-up storage unit 11 resets the on-line storage unit 10 to switch the back-up storage unit 11 to the on-line (step 142). Then, the storage unit 10 which has newly become on-line executes the processing of the next command (step 120).

A process when the time-out occurs is now explained in detail.

FIG. 8 illustrates a process of the time-out.

It is assumed here that the time-out 81 is indicated after the on-line storage unit 10 has processed two entries and the time-out 81 is indicated after the back-up storage unit 11 has processed three entries.

FIG. 9 illustrates a content of the resumption table 10-14 in the time-out case shown in FIG. 8. It is a table which is used when the execution result of the command by the on-line storage unit 10 indicates the time-out.

The resumption table 10-14 is a table to manage a resumption point of the back-up storage unit 11 by the on-line storage unit 10. When the execution result of the command indicates the time-out 81, an identifier of the corresponding command and the resumption points of the on-line and back-up storage units 10 and 11 are stored in the resumption table 10-14.

In the example of FIG. 8, the command identifier is "WRITE1", the resumption point of the on-line storage unit is "3" and the resumption point of the back-up storage unit is "4".

FIGS. 10A˜10C illustrate the response information of the command.

The response information for the time-out comprises a command identifier 90, response information 91 and a resumption point 92 as shown in FIG. 10A. In the example shown in FIG. 8, the identifier "WRITE1" indicating the command being executed is set in the command identifier 90, the "time-out" is set in the response information 91, "3" is set in the resumption position 92 of the on-line storage unit as shown in FIG. 10B and "4" is set in the resumption position of the back-up storage unit as shown in FIG. 10C.

FIG. 11 illustrates retry of the command.

When the time-out occurs in the on-line storage unit 10, the processing units 20˜22 must retransmit the command for which the time-out has occurred. The processing units 20˜22 set the resumption position in the command and issues the command to the on-line storage unit 10.

The on-line storage unit 10 stores all write data in the buffer 10-13 when it first receives the data.

Then, the on-line storage unit 10 looks up the resumption table 10-14 to continue the process from the point specified by the resumption point 92. The on-line storage unit 10 transfers the command to the back-up storage unit 11. When a plurality of write data are transmitted, it transmits all write data.

The back-up storage unit 11 stores all write data in the data buffer 10-13.

When the back-up storage unit 11 receives a retry command 290 from the on-line storage unit 10, it resumes the process from the resumption point specified by the resumption table 11-14. When the process is completed, the response message 291 is returned to the on-line storage unit 10.

In this manner, the on-line storage unit 10 receives the command from the processing units 20˜22 and transfers the received command to the back-up storage unit 11 so that the storage contents of the on-line storage unit 10 and the back-up storage unit 11 coincide.

When the status transition of the cache structure or the list structure occurs in the storage units 10 and 11, it should be notified to the processing units 20˜22. The status transition command is explained below.

FIG. 12 shows a coincidence control process of the status transition command.

The status transition command is a command to notify the occurrence of the status transition (writing or deletion of entry) in the list structure 40 (40-1, 40-1) or the cache structure 30 (30-0, 30-1) to the processing units 20˜22.

The status transition command is generated when the same event occurs in the on-line storage unit 10 and the back-up storage unit 11. Thus, whether the on-line and back-up storage units 10 and 11 operate normally or not may be checked by checking whether the status transition commands occurs synchronously in the on-line storage unit 10 and the back-up storage unit 11.

If the status transition occurs in only one of the storage units, the storage unit in which the status transition has not occurred is regarded as the occurrence of the fault and it is reset. This process is explained below.

When the status transition occurs in the on-line storage unit 10 (step 300), the on-line storage unit 10 issues the status transition command to the processing units 20˜22 (step 301). The on-line storage unit 10 check whether it has received the same status transition command from the back-up storage unit 11 or not within a predetermined period (step 302).

If the on-line storage unit 10 receives the status transition command, it is determined that the back-up storage unit 11 is normal, but if it does not receive the status transition command within the predetermined period (step 303), the back-up storage unit 11 is reset (step 304).

In this manner, the on-line storage unit 10 and the back-up storage unit 11 execute the received commands so that the storage contents coincide.

While the present invention has been specifically described in connection with the embodiment, it should be understood that the present invention is not limited to the above embodiment and various modifications may be made without departing from the scope of the invention.

In accordance with the present invention, the command is issued from the processing units to only the on-line storage unit and the command is transferred from the on-line storage unit to the back-up storage unit and the processing for the received commands is executed in the respective storage units. Thus, the overhead of the on-line storage unit is reduced when the duplex storage unit is structured by the hot stand-by system.

Further, even when a plurality of processing units, the on-line storage unit and the back-up storage unit are connected through the channel device, the storage contents of the on-line storage unit and the back-up storage unit are made coincident.

Claims (6)

What is claimed is:
1. In a computer system comprising two storage units having processors, one being an on-line storage unit and the other being a back-up storage unit, and a processing unit for accessing said storage units by issuing commands, a method for making storage contents of two storage units coincident comprising the steps of:
issuing a command from said processing unit to said on-line storage unit;
transferring the command received by said on-line storage unit to said back-up storage unit; and
executing respective processes for the received command in said on-line and back-up storage units.
2. A method for making storage contents of two storage units coincident according to claim 1 wherein said on-line storage unit transmits only a command to update a content of said on-line storage unit to said back-up storage unit.
3. A method for making storage contents of two storage units coincident according to claim 1 wherein when an execution result of the command by said on-line storage unit indicate normal completion, said on-line storage unit does not compare the execution result by said on-line storage unit with an execution result of the command by said back-up storage unit to said processing unit, and when the execution result of the command by said back-up storage unit indicates memory shortage or fault, the back-up storage unit is reset.
4. A method for making storage contents of two storage unit coincident according to claim 1 wherein when an execution result of the command by said on-line storage unit indicates normal completion, said on-line storage unit does not compare the execution result of said on-line storage unit with an execution result of the command by said back-up storage unit but transfers the execution result of the command by said on-line storage unit to said processing unit, and when the execution result of the command by said back-up storage unit indicates time-out, said on-line storage unit causes the processing for the command in said back-up storage unit to be executed to the last.
5. A method for making storage contents of two storage units coincident according to claim 1 wherein said processing unit is connected to said on-line storage unit and said back-up storage unit by a channel device.
6. A computer system comprising:
a first storage unit having a processor working as an on-line storage unit;
a second storage unit having another processor working as a back-up storage unit; and
a processing unit for accessing said first storage unit by issuing a command thereto; wherein
said first storage unit sends said command to said second storage unit and execute a process in accordance with said issued command, and said second storage unit executes a process in accordance with said sent command.
US08766455 1995-12-20 1996-12-12 Method for ensuring data coherency for redundat storage by transferring copies of access requests from on-line storage to back-up storage Expired - Fee Related US5845328A (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
JP33135695A JPH09171441A (en) 1995-12-20 1995-12-20 Storage matching method for duplex storage device and device therefor
JP7-331356 1995-12-20

Publications (1)

Publication Number Publication Date
US5845328A true US5845328A (en) 1998-12-01

Family

ID=18242773

Family Applications (1)

Application Number Title Priority Date Filing Date
US08766455 Expired - Fee Related US5845328A (en) 1995-12-20 1996-12-12 Method for ensuring data coherency for redundat storage by transferring copies of access requests from on-line storage to back-up storage

Country Status (2)

Country Link
US (1) US5845328A (en)
JP (1) JPH09171441A (en)

Cited By (24)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6167485A (en) * 1997-12-31 2000-12-26 Emc Corporation On-line data verification and repair in redundant storage systems
US6195761B1 (en) * 1997-12-31 2001-02-27 Emc Corporation Method and apparatus for identifying and repairing mismatched data
US6233696B1 (en) * 1997-12-31 2001-05-15 Emc Corporation Data verification and repair in redundant storage systems
US6393538B2 (en) 1998-09-18 2002-05-21 Fujitsu Limited Reproducing updates in a first file storage unit to a second file storage unit by transferring update information from the first unit to the second unit
US6408370B2 (en) * 1997-09-12 2002-06-18 Hitachi, Ltd. Storage system assuring data integrity and a synchronous remote data duplexing
US6526419B1 (en) 2000-06-09 2003-02-25 International Business Machines Corporation Method, system, and program for remote copy in an open systems environment
US20030046602A1 (en) * 2001-09-04 2003-03-06 Hitachi, Ltd. Data storage system
WO2003048936A1 (en) * 2001-11-29 2003-06-12 Agami Systems, Inc. Fault tolerance using logical checkpointing in computing systems
US20040064659A1 (en) * 2001-05-10 2004-04-01 Hitachi, Ltd. Storage apparatus system and method of data backup
US20040068628A1 (en) * 1997-09-12 2004-04-08 Hitachi, Ltd. Method and apparatus for data duplexing in storage unit system
US6732193B1 (en) 2000-06-09 2004-05-04 International Business Machines Corporation Method, system, and program for determining a number of write operations to execute
US6742093B2 (en) * 1998-02-26 2004-05-25 Hitachi, Ltd. Subsystem and method of reorganizing multiplexed data
US20040103164A1 (en) * 1999-04-26 2004-05-27 Hideo Tabuchi Disk subsystems and their integrated system
US20040230756A1 (en) * 2001-02-28 2004-11-18 Hitachi. Ltd. Three data center adaptive remote copy
US20040268178A1 (en) * 2003-06-19 2004-12-30 Fredin Gerald J Method for recovering data from a redundant storage object
US20050022049A1 (en) * 2000-01-28 2005-01-27 Hitachi, Ltd. Subsystems and an integrated system of these subsystems
US20050021627A1 (en) * 1997-01-08 2005-01-27 Hitachi, Ltd. Adaptive remote copy in a heterogeneous environment
US20050120092A1 (en) * 1998-09-09 2005-06-02 Hitachi, Ltd. Remote copy for a storage controller with reduced data size
US20050120093A1 (en) * 2001-05-10 2005-06-02 Hitachi, Ltd. Remote copy for a storgae controller in a heterogeneous environment
US20050267923A1 (en) * 2000-03-31 2005-12-01 Hitachi, Ltd. Method for duplicating data of storage subsystem and data duplicating system
US7167962B2 (en) 1999-08-19 2007-01-23 Hitachi, Ltd. Remote copy for a storage controller with reduced data size
US7404005B1 (en) 2000-06-09 2008-07-22 International Business Machines Corporation Method, system, and program for selecting one of multiple paths to communicate with a device
US8024761B1 (en) * 1999-04-01 2011-09-20 Cox Communications, Inc. Method and apparatus of load sharing and fault tolerance in an interactive video distribution system
US20160070491A1 (en) * 2014-09-10 2016-03-10 Fujitsu Limited Information processor, computer-readable recording medium in which input/output control program is recorded, and method for controlling input/output

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4577272A (en) * 1983-06-27 1986-03-18 E-Systems, Inc. Fault tolerant and load sharing processing system
US5193154A (en) * 1987-07-10 1993-03-09 Hitachi, Ltd. Buffered peripheral system and method for backing up and retrieving data to and from backup memory device
US5226157A (en) * 1988-03-11 1993-07-06 Hitachi, Ltd. Backup control method and system in data processing system using identifiers for controlling block data transfer
US5313612A (en) * 1988-09-02 1994-05-17 Matsushita Electric Industrial Co., Ltd. Information recording and reproducing apparatus including both data and work optical disk drives for restoring data and commands after a malfunction
US5317739A (en) * 1992-03-30 1994-05-31 International Business Machines Corp. Method and apparatus for coupling data processing systems
US5537574A (en) * 1990-12-14 1996-07-16 International Business Machines Corporation Sysplex shared data coherency method
US5548711A (en) * 1993-08-26 1996-08-20 Emc Corporation Method and apparatus for fault tolerant fast writes through buffer dumping

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4577272A (en) * 1983-06-27 1986-03-18 E-Systems, Inc. Fault tolerant and load sharing processing system
US5193154A (en) * 1987-07-10 1993-03-09 Hitachi, Ltd. Buffered peripheral system and method for backing up and retrieving data to and from backup memory device
US5226157A (en) * 1988-03-11 1993-07-06 Hitachi, Ltd. Backup control method and system in data processing system using identifiers for controlling block data transfer
US5313612A (en) * 1988-09-02 1994-05-17 Matsushita Electric Industrial Co., Ltd. Information recording and reproducing apparatus including both data and work optical disk drives for restoring data and commands after a malfunction
US5537574A (en) * 1990-12-14 1996-07-16 International Business Machines Corporation Sysplex shared data coherency method
US5317739A (en) * 1992-03-30 1994-05-31 International Business Machines Corp. Method and apparatus for coupling data processing systems
US5548711A (en) * 1993-08-26 1996-08-20 Emc Corporation Method and apparatus for fault tolerant fast writes through buffer dumping

Cited By (49)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7114049B2 (en) 1997-01-08 2006-09-26 Hitachi, Ltd. Adaptive remote copy in a heterogeneous environment
US20050021627A1 (en) * 1997-01-08 2005-01-27 Hitachi, Ltd. Adaptive remote copy in a heterogeneous environment
US7191303B2 (en) 1997-09-12 2007-03-13 Hitachi, Ltd. Storage controller
US6615332B2 (en) 1997-09-12 2003-09-02 Hitachi, Ltd. Storage system assuring data integrity and asynchronous remote data duplexing
US7657707B2 (en) 1997-09-12 2010-02-02 Hitachi, Ltd. Storage controller
US6408370B2 (en) * 1997-09-12 2002-06-18 Hitachi, Ltd. Storage system assuring data integrity and a synchronous remote data duplexing
US20040068628A1 (en) * 1997-09-12 2004-04-08 Hitachi, Ltd. Method and apparatus for data duplexing in storage unit system
US6988176B2 (en) 1997-09-12 2006-01-17 Hitachi, Ltd. Method and apparatus for data duplexing in storage unit system
US20040199737A1 (en) * 1997-09-12 2004-10-07 Akira Yamamoto Storage controller
US6389511B1 (en) 1997-12-31 2002-05-14 Emc Corporation On-line data verification and repair in redundant storage system
US6233696B1 (en) * 1997-12-31 2001-05-15 Emc Corporation Data verification and repair in redundant storage systems
US6167485A (en) * 1997-12-31 2000-12-26 Emc Corporation On-line data verification and repair in redundant storage systems
US6195761B1 (en) * 1997-12-31 2001-02-27 Emc Corporation Method and apparatus for identifying and repairing mismatched data
US6535994B1 (en) 1997-12-31 2003-03-18 Emc Corporation Method and apparatus for identifying and repairing mismatched data
EP0939371A3 (en) * 1998-02-26 2006-08-02 Hitachi, Ltd. Memory subsystem and method of reorganising multiplexed data
US6742093B2 (en) * 1998-02-26 2004-05-25 Hitachi, Ltd. Subsystem and method of reorganizing multiplexed data
US6993636B2 (en) 1998-02-26 2006-01-31 Hitachi, Ltd. Subsystem and method of organizing multiplexed data
US20050120092A1 (en) * 1998-09-09 2005-06-02 Hitachi, Ltd. Remote copy for a storage controller with reduced data size
US7246214B2 (en) 1998-09-09 2007-07-17 Hitachi, Ltd. Remote copy for a storage controller with reduced data size
US6393538B2 (en) 1998-09-18 2002-05-21 Fujitsu Limited Reproducing updates in a first file storage unit to a second file storage unit by transferring update information from the first unit to the second unit
US8024761B1 (en) * 1999-04-01 2011-09-20 Cox Communications, Inc. Method and apparatus of load sharing and fault tolerance in an interactive video distribution system
US20040103164A1 (en) * 1999-04-26 2004-05-27 Hideo Tabuchi Disk subsystems and their integrated system
US7373390B2 (en) 1999-04-26 2008-05-13 Hitachi, Ltd. Disk subsystems and their integrated system
US7167902B1 (en) 1999-04-26 2007-01-23 Hitachi, Ltd. Disk subsystems and their integrated system
US7167962B2 (en) 1999-08-19 2007-01-23 Hitachi, Ltd. Remote copy for a storage controller with reduced data size
US7130959B2 (en) 2000-01-28 2006-10-31 Hitachi, Ltd. Subsystems and an integrated system of these subsystems
US20050022049A1 (en) * 2000-01-28 2005-01-27 Hitachi, Ltd. Subsystems and an integrated system of these subsystems
US20050267923A1 (en) * 2000-03-31 2005-12-01 Hitachi, Ltd. Method for duplicating data of storage subsystem and data duplicating system
US6981008B2 (en) 2000-03-31 2005-12-27 Hitachi, Ltd. Method for duplicating data of storage subsystem and data duplicating system
US6732193B1 (en) 2000-06-09 2004-05-04 International Business Machines Corporation Method, system, and program for determining a number of write operations to execute
US8037171B2 (en) 2000-06-09 2011-10-11 International Business Machines Corporation System and program for selecting one of multiple paths to communicate with a device
US7404005B1 (en) 2000-06-09 2008-07-22 International Business Machines Corporation Method, system, and program for selecting one of multiple paths to communicate with a device
US6526419B1 (en) 2000-06-09 2003-02-25 International Business Machines Corporation Method, system, and program for remote copy in an open systems environment
US20080205301A1 (en) * 2000-06-09 2008-08-28 International Business Machines Corporation System and program for selecting one of multiple paths to communicate with a device
US20040230756A1 (en) * 2001-02-28 2004-11-18 Hitachi. Ltd. Three data center adaptive remote copy
US7194590B2 (en) 2001-02-28 2007-03-20 Hitachi, Ltd. Three data center adaptive remote copy
US20040064659A1 (en) * 2001-05-10 2004-04-01 Hitachi, Ltd. Storage apparatus system and method of data backup
US20050120093A1 (en) * 2001-05-10 2005-06-02 Hitachi, Ltd. Remote copy for a storgae controller in a heterogeneous environment
US7143252B2 (en) * 2001-05-10 2006-11-28 Hitachi, Ltd. Storage apparatus system and method of data backup
US7213114B2 (en) 2001-05-10 2007-05-01 Hitachi, Ltd. Remote copy for a storage controller in a heterogeneous environment
US7082506B2 (en) 2001-08-08 2006-07-25 Hitachi, Ltd. Remote copy control method, storage sub-system with the method, and large area data storage system using them
US6907543B2 (en) * 2001-09-04 2005-06-14 Hitachi, Ltd. Data storage system
US20030046602A1 (en) * 2001-09-04 2003-03-06 Hitachi, Ltd. Data storage system
US6954877B2 (en) 2001-11-29 2005-10-11 Agami Systems, Inc. Fault tolerance using logical checkpointing in computing systems
US20040199812A1 (en) * 2001-11-29 2004-10-07 Earl William J. Fault tolerance using logical checkpointing in computing systems
WO2003048936A1 (en) * 2001-11-29 2003-06-12 Agami Systems, Inc. Fault tolerance using logical checkpointing in computing systems
US20040268178A1 (en) * 2003-06-19 2004-12-30 Fredin Gerald J Method for recovering data from a redundant storage object
US7237141B2 (en) * 2003-06-19 2007-06-26 Lsi Corporation Method for recovering data from a redundant storage object
US20160070491A1 (en) * 2014-09-10 2016-03-10 Fujitsu Limited Information processor, computer-readable recording medium in which input/output control program is recorded, and method for controlling input/output

Also Published As

Publication number Publication date Type
JPH09171441A (en) 1997-06-30 application

Similar Documents

Publication Publication Date Title
US5276684A (en) High performance I/O processor
US4674036A (en) Duplex controller synchronization circuit for processors which utilizes an address input
US7117327B2 (en) Data processing system
US5530902A (en) Data packet switching system having DMA controller, service arbiter, buffer type managers, and buffer managers for managing data transfer to provide less processor intervention
US5418966A (en) Updating replicated objects in a plurality of memory partitions
US6615319B2 (en) Distributed mechanism for resolving cache coherence conflicts in a multi-node computer architecture
US5924119A (en) Consistent packet switched memory bus for shared memory multiprocessors
US6249520B1 (en) High-performance non-blocking switch with multiple channel ordering constraints
US6161198A (en) System for providing transaction indivisibility in a transaction processing system upon recovery from a host processor failure by monitoring source message sequencing
US5933653A (en) Method and apparatus for mirroring data in a remote data storage system
US6308284B1 (en) Method and apparatus for maintaining data coherency
US5878217A (en) Network controller for switching into DMA mode based on anticipated memory overflow and out of DMA mode when the host processor is available
US7246187B1 (en) Method and apparatus for controlling exclusive access to a shared resource in a data storage system
US6487202B1 (en) Method and apparatus for maximizing memory throughput
US7269666B1 (en) Memory utilization in a network interface
US4894828A (en) Multiple sup swap mechanism
US5659696A (en) Method and apparatus for determining address location and taking one of two actions depending on the type of read/write data transfer required
US4672537A (en) Data error detection and device controller failure detection in an input/output system
US20030167419A1 (en) Remote data mirroring system having a remote link adapter
US4604710A (en) System for converting data processing information to text processing format and vice versa
US4623997A (en) Coherent interface with wraparound receive and transmit memories
US6487643B1 (en) Method and apparatus for preventing starvation in a multi-node architecture
US20080109569A1 (en) Remote DMA systems and methods for supporting synchronization of distributed processes in a multi-processor system using collective operations
US5941964A (en) Bridge buffer management by bridge interception of synchronization events
US5829047A (en) Backup memory for reliable operation

Legal Events

Date Code Title Description
AS Assignment

Owner name: HITACHI, LTD., JAPAN

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:MAYA, YUZURU;ISONO, SOICHI;OHTSUJI, AKIRA;REEL/FRAME:008341/0322

Effective date: 19961202

FPAY Fee payment

Year of fee payment: 4

REMI Maintenance fee reminder mailed
REMI Maintenance fee reminder mailed
LAPS Lapse for failure to pay maintenance fees
FP Expired due to failure to pay maintenance fee

Effective date: 20061201